1 State Key Laboratory of Materials for Integrated Circuits, Shanghai Institute of Microsystem and Information Technology, Shanghai 200050, China; 2 University of the Chinese Academy of Sciences, Beijing 100049, China
Abstract The integration of artificial intelligence (AI) with satellite technology is ushering in a new era of space exploration, with small satellites playing a pivotal role in advancing this field. However, the deployment of machine learning (ML) models in space faces distinct challenges, such as single event upsets (SEUs), which are triggered by space radiation and can corrupt the outputs of neural networks. To defend against this threat, we investigate laser-based fault injection techniques on 55-nm SRAM cells, aiming to explore the impact of SEUs on neural network performance. In this paper, we propose a novel solution in the form of Bin-DNCNN, a binary neural network (BNN)-based model that significantly enhances robustness to radiation-induced faults. We conduct experiments to evaluate the denoising effectiveness of different neural network architectures, comparing their resilience to weight errors before and after fault injections. Our experimental results demonstrate that binary neural networks (BNNs) exhibit superior robustness to weight errors compared to traditional deep neural networks (DNNs), making them a promising candidate for spaceborne AI applications.
Corresponding Authors:
Dawei Bi
E-mail: davidb@mail.sim.ac.cn
Cite this article:
Qing Liu(刘清), Haomiao Cheng(程浩淼), Xiang Yao(姚骧), Zhengxuan Zhang(张正选), Zhiyuan Hu(胡志远), and Dawei Bi(毕大炜) Enhancing neural network robustness: Laser fault injection resistance in 55-nm SRAM for space applications 2025 Chin. Phys. B 34 046104
[1] Brunetti G, Campiti G, TaglienteMand Ciminelli C 2024 IEEE Access 12 76478 [2] Dario Izzo, Marcus Märtens and Pan B F 2019 Astrodynamics 3 287 [3] Benoot W 2023 European Data Handling & Data Processing Conference (EDHPC) pp. 1-6 [4] Li J, Yang Z and Luo Y 2024 Advances in Space Research 75 2184 [5] Rech P 2024 IEEE Transactions on Nuclear Science 71 377 [6] Brewer R M, Moran S L, Cox J, Sierawski B D, McCurdy MW, Zhang E X, Iyer S S,Schrimpf R D, Alles M L and Reed R A 2020 IEEE Transactions on Nuclear Science 67 108-115 [7] KCHAOU A, YOUSSEF W E H and TOURKI R 2017 International Journal of Advanced Computer Science and Applications 8 [8] Krcma M, Kotasek Z and Lojda J 2017 IEEE East-West Design & Test Symposium (EWDTS) pp. 1-6 [9] Das A and Touba N A 2018 Proceedings of the 2018 Great Lakes Symposium on VLSI GLSVLSI ’18 p. 219 [10] Libano F, Wilson B, Wirthlin M, Rech P and Brunhaver J 2020 IEEE Transactions on Nuclear Science 67 1478 [11] Rosales Vargas S A and García-Ortiz A 2024 13th International Conference on Modern Circuits and Systems Technologies (MOCAST) pp. 01-05 [12] Zhao X, Du X, Xiong X, Ma C, Yang W, Zheng B and Zhou C 2024 Chin. Phys. B 33 078501 [13] Eslami M, Ghavami B, Raji M and Mahani A 2020 Integration 71 154 [14] Condia J E R, Guerrero-Balaguera J D, Dos Santos F F, ReordaMS and Rech P 2022 IEEE International Test Conference (ITC) pp. 278-287 [15] Salami B, Onural E B, Yuksel I E, Koc F, Ergin O, Cristal Kestelman A, Unsal O, Sarbazi-Azad H and Mutlu O 2020 50th Annual IEEE/IFIP International Conference on Dependable Systems and Networks (DSN) pp. 138-149 [16] Liu W, Chang C H, Zhang F and Lou X 2020 57th ACM/IEEE Design Automation Conference (DAC) pp. 1-6 [17] Selmke B, Brummer S, Heyszl J and Sigl G 2016 Smart Card Research and Advanced Applications, eds. Homma N and Medwed M pp. 193- 205 [18] Hou X, Breier J, Jap D, Ma L, Bhasin S and Liu Y 2020 IEEE International Symposium on the Physical and Failure Analysis of Integrated Circuits (IPFA) pp. 1-6 [19] Yuan C and Agaian S S 2023 Artif. Intell. Rev. 56 12949 [20] Hubara I, Courbariaux M, Soudry D, El-Yaniv R and Bengio Y 2016 Proceedings of the 30th International Conference on Neural Information Processing Systems NIPS’16 p. 4114 [21] Rastegari M, Ordonez V, Redmon J and Farhadi A 2016 Computer Vision - ECCV 2016 pp 525-542 [22] Liu Z, Shen Z, Savvides M and Cheng K T 2020 Computer Vision - ECCV 2020 pp. 143-159 [23] Cai Y, Zheng Y, Lin J, Yuan X, Zhang Y and Wang H 2023 arXiv: 2305.10299v3cs.CV] [24] Huang J and Han J 2005 Sci. China-Phys., Mech. Astron. 48 113 [25] Zhang H, Liu Z, Xie Y, Jia Y and Zhang Z 2021 IEICE Electronics Express 18 20210404 [26] Zhang K, Zuo W, Chen Y, Meng D and Zhang L 2017 IEEE Transactions on Image Processing 26 3142 [27] Horé A and Ziou D 2010 2010 20th International Conference on Pattern Recognition pp. 2366-2369 [28] Schirmeier H, Borchert C and Spinczyk O 2015 45th Annual IEEE/IFIP International Conference on Dependable Systems and Networks pp. 319-330 [29] Deng Q, Shao C, Li H and Fang J 2019 IEEE 13th International Conference on Anti-counterfeiting, Security, and Identification (ASID) pp. 314-319 [30] Khoshavi N, Broyles C and Bi Y 2020 21st International Symposium on Quality Electronic Design (ISQED) pp. 99-104
Large energy-loss straggling of swift heavy ions in ultra-thin active silicon layers Zhang Zhan-Gang (张战刚), Liu Jie (刘杰), Hou Ming-Dong (侯明东), Sun You-Mei (孙友梅), Zhao Fa-Zhan (赵发展), Liu Gang (刘刚), Han Zheng-Sheng (韩郑生), Geng Chao (耿超), Liu Jian-De (刘建德), Xi Kai (习凯), Duan Jing-Lai (段敬来), Yao Hui-Jun (姚会军), Mo Dan (莫丹), Luo Jie (罗捷), Gu Song (古松), Liu Tian-Qi (刘天奇). Chin. Phys. B, 2013, 22(9): 096103.
No Suggested Reading articles found!
Viewed
Full text
Abstract
Cited
Altmetric
blogs
tweeters
Facebook pages
Wikipedia page
Google+ users
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.